server

Nope seems like a docker-compose issue

Try the following example.env :

CLEARML_SERVING_PORT=9090
CLEARML_WEB_HOST="http://<IP>:8080"
CLEARML_API_HOST="http://<IP>:8008"
CLEARML_FILES_HOST="http://<IP>:8081"

(I think the localhost is resolved to inside the container and not the host machine, hence the error)

I got only smth like this:

clearml-serving-triton        | I0701 08:32:58.580705 46 server.cc:250] Waiting for in-flight requests to complete.
clearml-serving-triton        | I0701 08:32:58.580710 46 server.cc:266] Timeout 30: Found 0 model versions that have in-flight inferences
clearml-serving-triton        | I0701 08:32:58.580713 46 server.cc:281] All models are stopped, unloading models
clearml-serving-triton        | I0701 08:32:58.580717 46 server.cc:288] Timeout 30: Found 0 live models and 0 in-flight non-inference requ

and my ~/clearml.conf

api { 
    web_server: 

    api_server: 

    files_server: 

    # test 3
    credentials {
        "access_key" = "91SFEX4BYUQ9YCZ9V6WP"
        "secret_key" = "4WTXT7tAW3R6tnSi8hzSKNjgkmgUoyv22lYT2FIzIfLoeGERRO"
    }
} 

What are you getting with:

curl http://<ip>:8008/auth.login

oh, I see one error, let me check fast

so it works with

It throws the same error

but it actually looks ok

I have to step away for a couple of hours
please let me know if you find something wrong

that's strange, maybe you should upgrade it

how d you start a docker-compose?

docker-compose --env-file example.env -f docker-compose.yml up -d

Did you get the same as well?

yeah, I tried the following
None
but haven't managed yet to make it work

serving

I changed port here:

clearml-serving-inference:
    image: allegroai/clearml-serving-inference:latest
    container_name: clearml-serving-inference
    restart: unless-stopped
    ports:
      - "9090:8080"

the way above works for me

Okay this seems correct...
Can you share both yaml files (server & serving) and env file?

I tried that, it didn't work. I was confused by the separate port parameter:

CLEARML_SERVING_PORT: ${CLEARML_SERVING_PORT:-8080}

which is only one port related in docker-compose-triton.yml
Can I test /auth.login somehow independently? Using curl or any other way. Which address does it suppose to have and which creds should I use?

maybe I'm missing something with credentials?

yeah, ok
but it didn't

I don't thing WEB_HOST is important, but what about FILE_HOST?
do I need to change it accordingly?

I haven't followed it so closely, but let me check

Yep this is fine

does it work for you?

doesn't work anyway

except access_key of course, they should be yours